KR20230052931A - Methods and systems for detecting skin contact reactions and recommending associated safe products - Google Patents

Abstract

The present application provides a system for detecting skin contact reactions to a set of allergens. During operation, the system allows a user to acquire an image of a patch area through a camera in a handheld device, where the patch area is placed on the patient's skin, and a patch containing a set of allergens is constant before the image is acquired. It is applied to the patched area for a period of time and then removed. Next, the system pre-processes the image so that the resulting pre-processed image of the patch area has a pre-specified size and orientation. The system then performs image processing operations on the preprocessed image to identify areas in the patch area that correspond to positive skin contact reactions. The system labels each identified area with the specific allergen that has been applied to the area and generates test results, which test results identify the specific allergen that caused positive skin contact reactions.

Description

Methods and systems for detecting skin contact reactions and recommending associated safe products

related application

35 USC. §119, U.S. Provisional Application Serial No. 63/067,256 entitled "Detection and Labeling of Skin Contact Reactions and Recommender System for Safe Product List" filed August 18, 2020 with inventor Ryan T. Lewinson et al. Priority is claimed, the contents of which are incorporated herein.

technology field

The disclosed embodiments relate generally to techniques for detecting skin allergic reactions. More specifically, the disclosed embodiments provide a system for detecting and labeling skin contact reactions and using this information to recommend associated safe products.

Approximately 20% of the general population experiences a rash caused by a skin allergen, a condition known as allergic contact dermatitis (ACD). ACD is a chronic condition that produces erythematous (red), pruritic (itching), painful, eczematous rashes, sometimes also associated with skin blisters or fissures (fissures). ACD is the result of a delayed hypersensitivity reaction to an allergen, which usually occurs 48 to 72 hours after exposure. These allergens can be found in the workplace or at home, in cosmetics or personal care products. ACD often affects the face, particularly the eyelids, along with the hands and feet; ACD can occur anywhere on the body as a result of allergen exposure. The most common allergens implicated in the development of ACD are metals (eg nickel, cobalt), fragrances, preservatives (eg methylisothiazolinone/methylchloroisothiazolinone, formaldehyde), topical antibiotics (eg eg neomycin, bacitracin) and dyes (eg p-phenylenediamine). ACD is a chronic disease with no cure, but can be avoided once the offending allergen is identified and removed from use.

Currently, the gold standard for diagnosing ACD is the “patch test,” a procedure that has been in existence for over 100 years. A patch test involves having a series of patches containing individual allergens suspended in petroleum jelly (or other medium) applied to the skin, usually the upper back, for approximately 48 hours. The consequences of this exposure (delayed hypersensitivity response) are evaluated after a further 48-72 hours. The map of the applied patch is then used as a reference to determine the individual's specific allergen.

Although patch testing has been used for many years, it has its drawbacks. For example, patch tests are typically only offered by dermatologists with expertise or a special interest in ACD, and such dermatologists are typically only conducted in large urban centers. As such, access to treatment is significantly limited for patients living in rural areas, and waiting times for all patients often exceed one year. During this long waiting time, the patient suffers. Anxiety regarding skin symptoms, visual changes in appearance, and their personal care products exacerbates patient morbidity. High operating costs and costs to patients are likely exacerbated by routinely screening each patient for a large number of allergens (typically >100), which inevitably includes allergens rarely found in common commercial products . Additionally, the overall time required to perform a patch test procedure makes most dermatologists rarely offer the procedure during their busy practice. Treatment of ACD ultimately requires identification and elimination of the causative allergen. However, limited access to health care providers, long waiting lists, geographic restrictions, and prohibitive costs all contribute to reduced access to care and increased patient morbidity.

Artificial intelligence-based decision-making has been growing rapidly in dermatology and skin care, with previous use cases being developed for detection of melanoma skin cancer, tracking skin features over time, and beauty advice. While these modern technologies have not yet been applied to patch testing, they offer significant opportunities to expand ACD testing by allowing patients to remotely and safely assess their response without physically seeing an experienced assessor, such as a dermatologist.

Accordingly, what is needed is a new technique for diagnosing ACD and identifying associated allergens without the drawbacks of existing patch test techniques.

The disclosed embodiments relate to a system for detecting skin contact reactions to a set of allergens. During operation, the system allows a user to acquire an image of a patch area through a camera in a handheld device, where the patch area is placed on the patient's skin, and a patch containing a set of allergens is constant before the image is acquired. It is applied to the patched area for a period of time and then removed. Next, the system pre-processes the image so that the resulting pre-processed image of the patch area has a pre-specified size and orientation. The system then performs image processing operations on the preprocessed image to identify areas in the patch area that correspond to positive skin contact reactions. Finally, the system labels each identified area with the specific allergen that has been applied to the area and generates test results, which test results identify the specific allergen that caused positive skin contact reactions.

In some embodiments, the patch was applied to the patch area for approximately 2 days and removed for approximately 2-3 days before an image was acquired.

In some embodiments, while allowing the user to acquire an image through the camera, the system displays to the user an image of what is being seen by the camera, and the user adjusts the position of the camera so that the entire patch area is visible. image can be obtained.

In some embodiments, more than one image is acquired and processed to account for different lighting conditions, skin tone, skin pigment, and skin texture variations.

In some embodiments, multiple images are acquired at different times. For example, the first image may be acquired at 48 hours and the second image may be obtained at 72 hours. It evaluates irritation versus allergic response based on the difference between images/viewpoints, and can also use both images for correction of orientation/registration markings, possibly based on background skin and other identified characteristics of the patches/markings. let it be

In some embodiments, preprocessing the image involves performing one or more of the following preprocessing operations: shifting the image; Rotate the image; scaling the image; cropping the image; adjusting the image for a baseline skin tone or color; enhancing or normalizing selected qualities of images using machine learning; and adjusting the brightness and/or contrast of the image.

In some embodiments, preprocessing the image includes normalizing the image using machine learning techniques to vary light conditions, skin tone, and contrast.

In some embodiments, one or more of the preprocessing operations are performed with reference to patch registration marks that were transferred from the patch to the skin while the patch was applied to the skin.

In some embodiments, while performing image processing operations to identify areas of the patch area that correspond to positive skin contact reactions, the system uses a machine learning model based on a change in skin color and/or skin texture to generate positive results. Recognize skin contact reactions.

In some embodiments, the machine learning model includes a neural network or related deep learning architecture.

In some embodiments, the system further uses the test results to generate a set of safe product recommendations for the patient, filtering the product catalog based on lists of known product ingredients.

In some embodiments, while generating a set of safe product recommendations, the system may use the patient's age, patient's gender, patient's skin type and tissue, patient's existing skin concerns, climate-related aspects of the patient's variable/selected region, and patient's Additional consideration is given to patient data and preferences, which may include product preferences.

In some embodiments, the secure set of product recommendations creates a personalized list that is also tailored and/or filtered for individual retailers or brands.

In some embodiments, pre-processing operations and image processing operations may be performed on a portable device; and a remote cloud computing system.

In some embodiments, the portable device includes one of a smart phone, a tablet computer and a digital camera with a processor.

The disclosed embodiments also relate to a patch that allows testing of a set of allergens for skin contact reactions. The patch includes a set of wells configured to hold a set of allergens. It also includes adhesive sections positioned on the patch to bond the patch to the patient's skin. Finally, the patch includes registration marks composed of dye or other temporary tattoos that are injected into the patch, the registration marks being transferred to the patient's skin when the patch is applied to the patient. (Note that registration markings appear on the patient's skin either on the patch itself or as a remnant after the patch is applied.)

In some embodiments, the patch further includes a water repellent coating.

In some embodiments, the patch further includes a backing configured to be peeled off to expose the adhesive sections and registration marks prior to application of the patch to a patient.

In some embodiments, a set of wells within a patch are shipped pre-filled with a set of allergens.

In some embodiments, the user loads the allergen set into the well set before the patch is applied to the patient.

In some embodiments, patch orientation registration markings are drawn on the skin by a patient or other observer, which may involve either tracing the periphery of the patch, or tracing within a pre-cut registration template. there is.

In some embodiments, the patch and allergen orientation is a previous photograph of the patient wearing the patch - this may include a goniometer-based measurement from a handheld device, a QR code reading from the patch surface, or manual labeling by the patient or observer. - is determined from

In some embodiments, patch orientation is learned from a variety of different well shapes and arrangements within the patch.

In some embodiments, patch orientation is determined from markings on the secondary adhesive layer that remain on the skin for a period of time after the patch is removed.

In some embodiments, patch orientation is determined manually.

1A and 1B present detailed flow diagrams depicting an exemplary process for detecting and labeling skin contact reactions and using this information to recommend associated safe products in accordance with disclosed embodiments.
2A and 2B present a flow chart depicting an exemplary process of how an image of a patch test result may be used to identify an allergen causing skin contact reactions in accordance with disclosed embodiments.
3 depicts an exemplary product recommendation system in accordance with disclosed embodiments.
4 shows an exemplary patch with multiple wells and associated registration marks for different allergens according to disclosed embodiments.
5 presents a high level flow diagram depicting a process for detecting skin contact reactions in accordance with disclosed embodiments.

The following description is presented to enable any person skilled in the art to make and use the present embodiments, and is presented in the context of a particular application and its requirements. Various modifications to the disclosed embodiments will be apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments and applications without departing from the spirit and scope of the present embodiments. Thus, the present embodiments are not limited to the presented embodiments, but are to be accorded the widest scope consistent with the principles and features disclosed herein.

The data structures and code described in this detailed description are typically stored on a computer-readable storage medium, which can be any device or medium capable of storing code and/or data for use by a computer system. Computer-readable storage media include volatile memory, non-volatile memory, magnetic and optical storage devices such as disk drives, magnetic tape, compact disc (CD), digital versatile disc or digital video disc (DVD), or Other media capable of storing computer readable media include, but are not limited to.

The methods and processes described in the Detailed Description section may be embodied as code and/or data, which may be stored on a computer readable storage medium as described above. When a computer system reads and executes code and/or data stored on a computer-readable storage medium, the computer system performs methods and processes embodied as data structures and code and stored within the computer-readable storage medium. Additionally, the methods and processes described below may be included in a hardware module. For example, a hardware module may include, but is not limited to, application specific integrated circuit (ASIC) chips, field programmable gate arrays (FPGAs), and other programmable logic devices now known or later developed. When a hardware module is activated, the hardware module performs methods and processes contained within the hardware module.

Argument

The present invention relates to computer-assisted methods for diagnosing and labeling skin contact allergens and irritation reactions, and using this information to recommend associated safe products. 1A and 1B show a flow diagram of a general approach. Referring to FIG. 1A , step 101 involves removing the patch test from the skin. A typical patch test procedure involves applying an array of common allergens and irritants in various concentrations to the skin and leaving them in place for a period of time (~48 hours). This array can be published standard or customized. Time is provided to allow sufficient skin exposure of the applied chemical to induce a skin response in sensitive individuals. Importantly, this reaction is safe and not life threatening. After removing the patch, an additional period of time (˜48 to 72 hours) is given before proceeding to the next step.

Step 102 involves the collection of photographs of the entire patch test area (or area to be analyzed). This can be realized with a handheld smart device, personal camera, or any other photographic device that transfers images to electronic media. This image is then displayed to the user in step 103.

Step 104 determines whether the entire area to be evaluated is within view on the image. This can be done manually through user decisions, or alternatively can be done electronically by computer aided determination of image quality. If the image quality is insufficient, the user is prompted to provide a new image; If not, the technique proceeds.

Step 105 involves image pre-processing which may or may not be required depending on the image source and image quality. These processing steps may include, but are not limited to, image cropping, image transformation, image rotation, and image brightness and contrast adjustment.

Step 106 passes the patch test image to a pre-trained artificial intelligence/machine learning model, which is used in step 107 to identify regions corresponding to positive responses. This machine learning model can be implemented using any of a number of convolutional neural networks, feedforward neural networks, object detection systems, or other machine learning models. By using an appropriately trained machine learning model, the system, once deployed, provides a rapid solution for analysis of image quality, patch orientation, reaction severity, baseline skin color, baseline skin tissue, and other complex features and /or can account for changes in variables. In the disclosed embodiment, the system implements a modified You Only Look Once v2 (YOLOv2) technique, where the image is rectified by convolution, batch normalization, rectified linear unit activation, in which feature weights are learned. It goes through a series of layers with linear unit activation and max pooling. Using the trained model, features can be extracted from a given layer to enable final object detection. In this way, objects (ie, positive reactions) can be detected and marked by bounding boxes, and associated coordinates relative to the original image can be stored. This process is illustrated in the diagrams shown in FIGS. 2A and 2B. This information may be used at step 118 to enable data interpretation and associated patient recommendation if the user decides not to use the actions performed at steps 112-117.

Data sets 108 and 109 provide supplemental information identifying the specific chemicals tested and the specific patch layout for the chemicals. A user can manually enter this information, or the information can be automatically imported from data sets 108 and 109. In one embodiment, a QR code associated with an applied patch may be scanned to retrieve this data. In another embodiment, a list of predefined patch layouts may be displayed and selected by the user.

Step 110 integrates the data from the machine learning output from step 107 and the supplemental patch layout information from data sets 108 and 109 to label positive reactions according to the chemical corresponding to that reaction site. This information is then displayed to the user at step 111 as a bounding box overlay corresponding to the original image and positive patch results, which can then be labeled according to the specific contact allergen or irritant causing the positive reactions (FIG. 2A and FIG. see 2b).

Step 118 can be initiated directly if the user wishes to manually interpret this information. However, as shown in boxes 112-117, the present invention also combines patch test result data with product information from product catalog 114 and user data from data set 116 and user preferences to provide product recommendations. Provides a technique for integrating with the same external data. (See Figure 3 for more details.) This can be important because some of the chemicals used in patch testing found in products are not familiar to the general public and clinicians without patch testing experience.

The labeled images and patch test results (ie, allergens and irritants that cause positive reactions) from data set 112 are exported from the machine learning system as stored data. Depending on the platform used to initiate this process, the data may be stored locally on the user's smart device, locally on the computer system, or on a remote cloud-based platform.

The data is then passed to the product filtering mechanism at step 113, which includes external data from the product catalog 114. Product catalog 114 includes commercially available products (personal care products, cosmetics, cosmeceuticals, general purpose/cleansing, occupational), as well as common products known to cause contact dermatitis (e.g., detergents). , household cleaners, etc.) Product catalog 114 includes data corresponding to ingredient lists (including allergens and irritants), product intended use (eg, moisturizers, sunscreens, shampoos, detergents, etc.), customer ratings, and prices, but hereby Not limited. By integrating known product ingredient lists from product catalog 114 with data from patch test results, data in product catalog 114 is filtered to remove all products containing allergens and irritants that tested positive in step 115. It can be. As new products are introduced to the market frequently, this list will be updated in real time.

Data set 116 includes available user data and preferences. For example, this data may include patient age and sex; skin type; pigments and textures (such as oily/dry skin, etc.) obtained through user input or through image classification; geography (eg, related to climate, temperature, humidity, etc.); other skin issues; and information about other products used, but is not limited thereto. By integrating this data with data from safe product list 115, user-specific product recommendations can be created that exclude products containing chemicals that elicited positive patch test responses.

This user-specific product recommendation can be generated using a simple “if this, then that” (IFTTT) technique, or a machine learning approach can be used instead, where the feature weights are used to train the recommender system to produce the desired outputs. running by doing A clustering model can also be developed to identify groups of individuals with data similarities. These "smart" approaches using machine learning and clustering analysis can identify correlations that are not readily apparent to help deal with multivariate complexity.

For example, a 27-year-old female with a Fitzpatrick skin phototype II (fair skin) and a history of atopic dermatitis (eczema) living in Calgary, Canada (usually low-temperature dry) found that, by IFTTT or machine learning, she was positive. It would be recommended to use a moisturizing cream (considering her lighter skin phototype) that excludes any of the contact allergens and irritants that can be tested for can, and the technique can also provide a list of products that meet these criteria. Using a machine learning approach, it can also be determined that individuals with similar geography, age, gender, and/or other variables tend to have higher personal preference scores for a particular product type, which can guide product recommendations. can help

Because of this, the system: (1) enables users to perform patch testing at home or in a clinical environment; (2) identifying benign allergens or irritants by artificial intelligence; (3) Computer-assisted smart recommendations for products that integrate user data and personal preferences to provide a guide for product selection to optimize skin care, as well as exclude allergens or irritants that have tested positive in patch tests. to provide.

Exemplary Applications

The system can be used with prefabricated consumer grade patches with common allergens for patch testing. These patches can be applied to users outside of healthcare facilities, and patch test allergen layouts can be determined via QR codes or in-app product selectors. Next, an image can be acquired by the user using a smart device with a camera on the day the patch test results are to be interpreted, and the image can be forwarded to a cloud-based network, which can be applied to the image to identify skin contact reactions. Apply a pre-trained machine learning model. By integrating the identified skin contact reactions with a defined patch test layout, the system identifies and labels allergens that cause benign skin reactions.

This data is then passed to a recommendation system, which filters a library of products based on relevant ingredient lists and removes products containing positive allergens. If the user has provided additional personal data via the user profile, this personal data can be used to refine the list of safe products by machine learning or IFTTT techniques to ensure safe, positive tested individuals are free of any contact allergens and their skin type and It is used to create a personalized list of recommended products matching your personal information and preferences. This list of recommended products will be fed back to the user's device stored in the associated user profile, with shortcut product links to facilitate purchases from the recommended product list that can be accessed and revisited over time. can

In another example, the process described above is repeated, but the user data and patch test result data are used by a machine learning system or other technique to enable manufacturing of a customized skin care product for the user.

In another example, an open application test is performed in which a user suspends a household product in petroleum jelly (Vaseline Jelly) and applies a patch containing petroleum jelly to the skin. This system is then used to determine if a positive reaction has occurred. Next, by using the product catalog, tested products can be cross-referenced (either manually or via automated computer-based technology) to identify likely offending allergens, and related products.

In a further example, the system is used by local/remote health care providers in locations where access to patch testing may be limited. A prefabricated patch test with a common allergen is applied to the patient, and then, on a return visit, the physician can use the system to store the image and associated positives in the patient's electronic medical record for later access and reference purposes, and , it is possible to generate a list of products that are safe for the patient. In this way, the system facilitates the initial diagnosis and management of allergic contact dermatitis, while the patient awaits later referral to a specialist center for further allergen testing if needed.

In a further example, the system may be used in specialized dermatology centers that offer patch testing, whereby the technology is used to document all positive reactions and quickly identify the causative allergen based on a user-defined patch test layout. do. These data can then be exported to the patient's electronic medical record for later reference, and an associated custom product list can be created. Rather than manually documenting 60 individual patch test results and then manually sorting and sequencing a list of safe products, this allows the patch test analysis procedure to be facilitated through computer-assisted technology, resulting in seeing more patients in a given period of time. make it possible

patch design

4 depicts an exemplary patch 400 that allows testing a set of allergens for skin contact reactions in accordance with disclosed embodiments. The patch 400 includes a set of wells 401-406 configured to hold a set of allergens A, B, C, D, E and F. Note that well sets 401-406 may be shipped pre-filled with allergen sets. Alternatively, a user may load a set of allergens into the well sets 401-406 before the patch is applied to the patient. The patch also includes a set of registration marks 411-414 composed of a temporary tattoo or a dye such as henna, such that when the patch is applied to the patient, the registration marks 411-414 are transferred to the skin of the patient. injected into me The patch may also include a water-repellent coating (not shown) to allow the patient to shower when wearing the patch. The patch may also include a backing (not shown) configured to be peeled off to expose the adhesive sections and registration marks prior to application of the patch to the patient. As mentioned in the subject matter section, there are many other alternative approaches to matching patch layouts.

Skin contact reaction detection

5 presents a high level flow diagram depicting a process for detecting skin contact reactions in accordance with disclosed embodiments. During operation, the system allows a user to acquire an image of a patch area through a camera in a handheld device, where the patch area is placed on the patient's skin, and a patch containing a set of allergens is constant before the image is acquired. It is applied to the patch area for a period of time and then removed (step 502). Next, the system pre-processes the image so that the resulting pre-processed image of the patch area has a pre-specified size and orientation (step 504). The system then performs image processing operations on the preprocessed image to identify areas in the patch area that correspond to positive skin contact reactions (step 506). Finally, the system labels each identified area with the specific allergen that was applied to the area, generating test results that identify the specific allergen that caused the positive skin contact reactions (step 506).

Various modifications to the disclosed embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments and applications without departing from the spirit and scope of the present invention. Accordingly, the present invention is not limited to the embodiments presented, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

The foregoing description of the embodiments has been presented for purposes of illustration and description. They are not intended to be exhaustive or to limit this description to the forms disclosed. Accordingly, many changes and modifications will be apparent to those skilled in the art. Also, the above disclosure is not intended to limit this description. The scope of this description is defined by the appended claims.

Claims (37)

As a method for detecting skin contact reactions to a set of allergens,
enabling a user to acquire an image of a patch area via a camera in a handheld device, wherein the patch area is placed on the patient's skin and a patch containing the set of allergens is placed for a period of time before the image is acquired. applied to the patch area and removed -;
pre-processing the image so that the resulting pre-processed image of the patch area has a pre-specified size and orientation;
performing image processing operations on the unprocessed or preprocessed image to identify areas in the patch area that correspond to positive skin contact reactions; and
labeling each identified area with a specific allergen that has been applied to the area to generate test results, wherein the test results identify the specific allergen that caused positive skin contact reactions. The method of claim 1 , wherein the patch was applied to the patch area for about 2 days before the image was acquired and removed for about 2 to 3 days. 2. The method of claim 1, wherein enabling the user to acquire the image via the camera:
displaying to a user an image of what is being viewed by the camera; and
And enabling the user to obtain a visible image of the entire patch area by adjusting the position of the camera. The method of claim 1 , wherein more than one image is acquired and processed to account for different lighting conditions, skin tone, and skin texture variations. The method of claim 1 , wherein multiple images are acquired at different times. The method of claim 1 , wherein preprocessing the image comprises the following preprocessing operations:
shifting the image;
rotating the image;
scaling the image;
cropping the image;
adjusting the brightness of the image;
adjusting the contrast of the image;
adjusting the image for a baseline skin tone, pigment or color; and
and performing one or more of the operations of enhancing or normalizing the selected quality of the image using machine learning. 7. The method of claim 6, wherein one or more of the preprocessing operations are performed with reference to patch registration marks that were transferred from the patch to the skin while the patch was applied to the skin. The method of claim 1 , wherein preprocessing the image includes normalizing the image using machine learning techniques to vary light conditions, skin tone, and contrast. The method of claim 1 , wherein performing image processing operations to identify the areas of the patch area corresponding to the positive skin contact responses uses a machine learning model based on changes in skin color/pigmentation and/or skin texture. Recognizing positive skin contact reactions using a method. 9. The method of claim 8, wherein the machine learning model comprises a neural network or related deep learning architecture. The method of claim 1 , further comprising filtering a product catalog based on lists of known product ingredients using the test results to create a set of safe product recommendations for the patient. 12. The method of claim 11, wherein generating a set of safe product recommendations comprises: patient age, patient gender, patient skin type and tissue, existing skin concerns, climate-related aspects of the patient's current or selected region, and the patient's product further entails taking into account patient data and preferences, which may include preferences. 13. The method of claim 12, wherein the set of secure product recommendations creates personalized listings that are also tailored and/or filtered for individual retailers or brands. The method of claim 1 , wherein the preprocessing operations and the image processing operations are performed on the portable device; and a remote cloud computing system. The method of claim 1 , wherein the portable device comprises one of a smart phone, a tablet computer and a digital camera with a processor. A system for detecting skin contact reactions to a set of allergens, comprising:
A portable device comprising a camera configured to allow a user to acquire an image of a patch area, wherein the patch area is placed on the skin of a patient and a patch containing the set of allergens is placed on the skin for a period of time before the image is acquired. Applied to the patch area and then removed ―; and
A processing mechanism comprising:
to pre-process the image such that the resulting pre-processed image of the patch area has a pre-specified size and orientation;
on the preprocessed image, to perform image processing operations to identify areas in the patch area that correspond to positive skin contact reactions; and
A system configured to label each identified area with a specific allergen that has been applied to that area to generate test results, wherein the test results identify the specific allergen that caused positive skin contact reactions. 17. The system of claim 16, wherein the patch was applied to the patched area for approximately 2 days and removed approximately 2-3 days before the image was acquired. 17. The method of claim 16, allowing the user to acquire the image via the camera, the system:
display to the user an image of what is being viewed by the camera; and
Wherein the system allows the user to obtain a visible image of the entire patch area by adjusting the position of the camera. 17. The system of claim 16, wherein more than one image is acquired and processed to account for different lighting conditions, skin tone, and skin texture variations. 20. The system of claim 19, wherein multiple images are acquired at different times. 17. The method of claim 16, wherein while preprocessing the image, the system performs the following preprocessing operations:
shifting the image;
rotating the image;
scaling the image;
cropping the image;
adjusting the brightness of the image;
adjusting the contrast of the image;
adjusting the image for a baseline skin tone or color; and
and performing one or more of enhancing or normalizing a selected quality of the image using machine learning. 22. The system of claim 21, wherein one or more of the preprocessing operations are performed with reference to patch registration marks that were transferred from the patch to the skin while the patch was applied to the skin. 17. The system of claim 16, wherein while preprocessing the image, the system normalizes the image using machine learning techniques to vary light conditions, skin tone, and contrast. 17. The method of claim 16, wherein while performing image processing operations to identify the areas of the patch area that correspond to the positive skin contact responses, the system performs a machine learning model based on changes in skin color and/or skin texture. To recognize positive skin contact reactions using a system. 25. The system of claim 24, wherein the machine learning model comprises a neural network or related deep learning architecture. 17. The system of claim 16, wherein the processing mechanism further uses the test results to generate a set of safe product recommendations for the patient, filtering a product catalog based on lists of known product ingredients. 27. The method of claim 26, wherein while generating the set of safe product recommendations, the processing mechanism determines the patient's age, patient's gender, patient's skin type and tissue, existing skin concerns, climate-related aspects of the patient's current or selected region, and the The system further considers patient data and preferences, which may include the patient's product preferences. 28. The system of claim 27, wherein the set of secure product recommendations creates personalized listings that are also tailored and/or filtered for individual retailers or brands. 17. The method of claim 16, wherein the processing mechanism comprises: the portable device; and a remote cloud computing system. 17. The method of claim 16, wherein the portable device comprises one of a smart phone, a tablet computer and a digital camera with a processor. A patch enabling testing of a set of allergens for skin contact reactions, comprising:
said patch comprising a set of wells configured to retain said set of allergens;
adhesive sections positioned on the patch to bond the patch to the patient's skin; and
The patch comprising registration marks that are transferred to the skin of the patient when the patch is applied to the patient. 32. The patch of claim 31, wherein the registration marks consist of dye or temporary tattoos that are injected into the patch. 32. The patch of claim 31, wherein registration marks consist of a secondary adhesive layer that remains on the skin for a period of time after the patch is removed. 32. The patch of claim 31, wherein the patch further comprises a water repellent coating. 32. The patch of claim 31, wherein the patch further comprises a backing configured to be peeled off to expose the adhesive sections and the registration marks prior to application of the patch to the patient. 32. The patch of claim 31, wherein a set of wells in the patch are shipped pre-filled with the set of allergens. 32. The patch of claim 31, wherein a user loads a set of allergens into the set of wells before the patch is applied to the patient.

Images